9.1 A 7nm 4-Core AI Chip with 25.6TFLOPS Hybrid FP8 Training, 102.4TOPS INT4 Inference and Workload-Aware Throttling
2021 IEEE International Solid- State Circuits Conference (ISSCC)（2021）
关键词
workload-aware throttling,low-precision computation,AI hardware accelerators,cloud,edge platforms,robust deep learning model,high-precision computation,power management,4-core AI chip,EUV technology,iso-accurate models,DLFIoat16,edge power-performance,product power envelope,TOPS INT4 inference,architecture optimization,aggressive circuit,cloud platforms,cutting-edge algorithmic
AI 理解论文
溯源树
样例
生成溯源树，研究论文发展脉络
Chat Paper
正在生成论文摘要